Journal Logo

Original Research

Gender Differences in Adolescent Marijuana Use and Associated Psychosocial Characteristics

Schepis, Ty S. PhD; Desai, Rani A. PhD; Cavallo, Dana A. PhD; Smith, Anne E. PhD; McFetridge, Amanda BS; Liss, Thomas B. BS; Potenza, Marc N. MD, PhD; Krishnan-Sarin, Suchitra PhD

Author Information
Journal of Addiction Medicine: March 2011 - Volume 5 - Issue 1 - p 65-73
doi: 10.1097/ADM.0b013e3181d8dc62
  • Free


Marijuana use among adolescents is recognized as a significant public health problem, with evidence indicating that its use produces significant neurobiological, psychosocial, and health consequences (Bray et al., 2000; Georgiades and Boyle, 2007; Harvey et al., 2007; Schneider, 2008). Estimates from 2 nationwide surveys, the Monitoring the Future (MTF) Surveys and the National Survey on Drug Use and Health, indicate that the prevalence of marijuana use among adolescents is exceeded only by the prevalence of alcohol or tobacco use (Johnston et al., 2008; Substance Abuse and Mental Health Services Administration, 2008). Although use rates have declined since the late 1990s, data from the MTF indicated that nearly 42% of high school seniors had used marijuana at some point in their lives, with >30% using in the past year and ∼20% in the past month (Johnston et al., 2008).

Given the relatively high rates of use, the risks associated with marijuana use by adolescents are particularly alarming. Adolescents who use marijuana regularly or heavily have higher levels of anxiety (Dorard et al., 2008), depressive symptoms (Medina et al., 2007), suicidality (Pedersen 2008) and externalizing behavior (Monshouwer et al., 2006) than nonusers. Marijuana use is also associated with an increased risk of developing psychotic symptoms in a dose-dependent fashion (Di Forti et al., 2007; Moore et al., 2007) and with an increased chance of the later development of depressive, bipolar, or anxiety diagnoses (Wittchen et al., 2007). Tobacco use (Georgiades and Boyle, 2007), nicotine dependence (Patton et al., 2005), and other substance use diagnoses (Wittchen et al., 2007) have been associated with adolescent marijuana use, and it may mediate the progression to heavier (eg, cocaine) substance use (Fergusson et al., 2008), though this is not universally found (Tarter et al., 2006).

Furthermore, heavier levels of marijuana use are associated with poorer sleep (Bolla et al., 2008), respiratory problems (Aldington et al., 2007; Brook et al., 2008), cancer (Berthiller et al., 2008), and a host of neurocognitive deficits, including attentional, learning, memory, and intellectual functioning decrements (Fried et al., 2002; Di Forti et al., 2007; Harvey et al., 2007; Brook et al., 2008). Finally, regular adolescent marijuana use is associated with poorer school performance (Brook et al., 2008; Leatherdale et al., 2008), deviant peer affiliation (Reboussin et al., 2007), and school dropout (Bray et al., 2000). The adverse profile associated with adolescent marijuana use is compounded by indications that use of marijuana persists into young adulthood and beyond for most adolescent users (Patton et al., 2007; Perkonigg et al., 2008). Together, the data indicate that understanding the characteristics of adolescent marijuana users is an important research and public health goal; such an understanding could be used to target at-risk individuals through prevention programs or early intervention.

One way to further this goal is to examine potential gender differences among adolescent marijuana users. Traditionally, female substance use and differences between male and female substance users have been understudied. Recent research has strongly indicated that male and female substance use are potentially different phenomena, with different motivations, use trajectories, consequences, barriers to treatment, and patterns of relapse to substance use following abstinence (Brady and Randall 1999; Walitzer and Dearing 2006). Examination of gender differences in adolescent marijuana use is a particularly understudied area, underscoring the need for further investigations. Indeed, initial work indicates that male and female adolescents differ in terms of their use rates, trajectories, and psychosocial correlates, among users.

For instance, males seem to be more likely to use marijuana, with the MTF indicating past year use among 29% of males and 24% of females; the National Survey on Drug Use and Health indicated that 17% of adolescent males and 15% of females had used marijuana during their lifetime (Johnston et al., 2008; Substance Abuse and Mental Health Services Administration 2008). Kandel and Chen (2000) found an earlier age of onset of marijuana use for males and found that males were more likely to be on a heavier use trajectory; it also seems that the risk factors for the onset of marijuana use differ somewhat by gender (Guxens et al., 2007). Furthermore, males seem to be more likely than females to become marijuana dependent in the first few years after initiation (Wagner and Anthony, 2007). Finally, Ridenour et al. (2006) found that females tended to have shorter, but nonsignificant, intervals between onset of marijuana use and either first problem or cannabis dependence.

However, few studies have examined current users to evaluate potential gender differences in demographic or psychosocial characteristics in current adolescent users of marijuana. Pedersen et al. (2001) examined the effects at 13 and 14 years of age of conduct problems on the initiation of marijuana use during an 18-month follow-up period. They found that baseline conduct problems promoted the likelihood of marijuana use initiation during the follow-up, with an especially pronounced effect in females. Furthermore, the type of conduct problems evidenced was important: for males, serious conduct problems were stronger promoters of initiation, but in females, covert or aggressive conduct problems were more robust.

Arguably, the most comprehensive study of gender differences in the correlates of adolescent marijuana use was conducted by Tu et al. (2008). They examined a sample of 8225 Canadian secondary school students participating in a cross-sectional survey. The authors found 3 gender differences: (1) grade level in school was predictive of male frequent marijuana use (defined as 3 to 9 episodes of use in the previous 30 days) but not of frequent use by females; (2) self-reported poor mental health increased the likelihood of either frequent or heavy use (10 or more episodes in the past 30 days) among females but not among males; and (3) Aboriginal ethnicity was associated with frequent or heavy use among males but not females. However, it should be noted that the authors did not test gender-based interactions, instead only noting when a factor was a significant main effect correlate in just 1 gender but not the other. Although this method may identify particularly robust predictors, it is not as statistically powerful as testing for interactions with gender, assuming the implicit assumptions of testing for interactions are met.

Thus, this study attempted to further examine the gender-related correlates of current marijuana use through the use of a survey of high-risk behavior participation in Connecticut high school students. The correlates of lifetime and past month marijuana use were evaluated separately by gender and then evaluated for potential differences by gender (ie, gender-by-correlate interactions) using χ2 and regression analyses. There were 2 primary aims of this study: (1) to evaluate which demographic, psychosocial, and risk behavior characteristics are associated with past month or lifetime marijuana use, separately for males and females and (2) to evaluate whether gender differences exist in the odds of marijuana use associated with the selected correlates. We hypothesized that adverse measures of health and functioning would be associated with marijuana use in both boys and girls. Although gender differences were expected in the strengths of the associations, given the lack of prior research, a priori hypotheses were not made about the nature and direction of these gender-related differences in the associations with marijuana use. On the basis of data indicating a more rapid progression from initial use to problematic use of abused substances in females when compared with males, we additionally hypothesized that girls would demonstrate a more rapid transition from initial to regular use of marijuana.


The procedures and sampling design of this survey have been described in detail in Schepis et al. (2008).

Study Procedures

All public high schools in Connecticut were invited to participate in the survey. To encourage participation, schools were offered a brief report detailing the prevalence of each risk behavior assessed. With participating schools, a passive consent procedure was developed. This procedure involved mailing letters to parents of all students in the school to inform them about the study and outline the procedure to deny permission for their child to take part in the study. In cases where a parent did not make contact to deny permission, permission was assumed. Before survey administration, a list of students who were denied permission to participate was compiled; those students were asked to quietly complete schoolwork while the survey was administered. All procedures were approved by the participating schools and the Institutional Review Board of the Yale University School of Medicine.

On the day of survey administration, research staff was on site to explain, distribute, and collect surveys and to answer any participant questions. All students were instructed that participation was fully voluntary and that they could refuse to participate at any time. All data were double entered, with random spot checks and examination of out-of-range responses to ensure accuracy. The sample obtained from this survey is demographically similar to the sample of Connecticut adolescents aged 14 to 18 years from the 2000 US Census.


The full survey was composed of 153 questions assessing demographics, substance use, and other risk behavior participation. Lifetime marijuana and past month marijuana use were the dependent variables. Potential correlates chosen fell into 3 domains: demographic, psychosocial, and risk behavior correlates. Demographic variables included race/ethnicity (African American, white, Hispanic/Latino, and Asian/Other), gender, who the participant lives with (both parents, 1 parent, or in another arrangement), average grades in school (As/Bs, Bs/Cs, or Cs and below), and grade in school (9th, 10th, 11th, or 12th). Participants were allowed to choose as many races/ethnicities as they felt applied to them. Psychosocial variables included whether the participant engages in extracurricular activities and whether the participant has a job. Finally, risk behaviors included past year gambling, past month cigarette use, past month alcohol use, past month binge alcohol use, lifetime nonmedical use of steroids, past year physical fighting, carrying a weapon in the past month, 2 or more weeks of depressed mood with anhedonia in the past year, and lifetime self-harm. Time from the age of initiation of marijuana use to age of regular use was examined using 2 items asking participants the age at which they began marijuana use and the age at which they began regular use, which was assessed by asking if participants considered themselves to be regular marijuana users. Answer choices for both items were 8 years or younger, 9 to 10 years, 11 to 12 years, 13 to 14 years, 15 to 16 years, or 17 years or older.

Data Analyses

Distribution characteristics of all variables were examined. Three sets of analyses were performed. First, χ2 analyses evaluated demographic, psychosocial, and risk behavior correlates separately by gender. Second, hierarchical logistic regressions were used to calculate adjusted odds ratios (AORs) for past month or lifetime marijuana use, given presence of a specific correlate. Grade in school was controlled for, as older students were significantly more likely to use marijuana. These analyses were stratified by gender to evaluate effects in males and females separately. Finally, hierarchical logistic regressions were used to evaluate potential gender interactions. Again, grade in school was controlled for through entry in the first block of variables. In the second block, the specific correlate and gender were entered. In the third, the interaction term for the correlate and gender were entered. When not specifically listed, the significance level for all analyses was set at a P value below 0.05. Finally, the analysis of time from initiation of marijuana use to initiation of regular use used a Mann-Whitney U analysis with gender as the independent variable and difference in age category (listed above) endorsed for age of initiation and age of regular use initiation; only adolescents who had transitioned to regular marijuana use were included in these analyses.


Demographic and Marijuana Use Characteristics

A total of 4523 adolescents participated in the survey, with 2124 males (47.0%) and 2345 females (51.8%). Fifty-four participants (1.2%) did not enter a gender and were not included in these analyses. Of participants, 1906 males (89.7% of males) and 2191 females (93.4% of females) completed the marijuana use section and were included in this study. Included participants had a mean age of 15.86 years (standard deviation = 1.26); 10.2% were African American, 14.0% were Hispanic/Latino, 18.3% were Asian or another ethnicity, and 75.8% were white. Participants were allowed to choose as many races/ethnicities as they felt applied to them, so data on racial/ethnic background will total to above 100%. Of participants with complete marijuana use data, 1655 participants (40.4%) endorsed lifetime marijuana use and 1004 participants (24.5%) endorsed past 30-day use (participants could be members of both the lifetime and past 30-day marijuana use groups); 2442 participants (59.6%) denied lifetime marijuana use. By gender, 854 females (39.0% of females) and 802 males (42.1% of males) endorsed lifetime marijuana use; 491 females (22.4% of females) and 513 males (26.9% of males) endorsed past 30-day use. Both of these are significant differences (lifetime: χ2 = 4.07, P = 0.044; past 30-day: χ2 = 11.10, P < 0.001).

Participants With Missing Data

The χ2 analyses revealed that participants excluded from analyses were more likely to be male (χ2 = 19.96, P < 0.001), African American (χ2 = 38.34, P < 0.001), and Hispanic/Latino (χ2 = 22.37, P < 0.001); white participants were more likely to be included (χ2 = 50.19, P < 0.001). There was a trend level difference for younger participants to be more likely to have complete data (F(1, 427.092) = 2.90, P = 0.089). Also, participants without sufficient data were more likely to have ever used cigarettes (χ2 = 37.48, P < 0.001) or alcohol (χ2 = 7.62, P = 0.007) and were more likely to have used alcohol in a binge fashion (χ2 = 5.78, P = 0.016) in the past 30 days. No differences between were found in lifetime self-harm (χ2 = 2.70, P = 0.109) or in taking part in a physical fight in the past year (χ2 = 1.55, P = 0.213).

Characteristics Associated With Lifetime Marijuana Use, Stratified by Gender

For both male and female participants, self-reported grade in school, average grades, and family structure were significant correlates. Across gender, ninth graders were significantly less likely than 11th or 12th graders to have used marijuana, and those with an A/B average in school were less likely to have used than those with a B/C average or Cs and below. Males and females differed slightly in the association of family structure and lifetime marijuana use: males in a 2-parent household were less likely to have ever used than males in either a 1-parent household or some other household (eg, with grandparents) and for females, those in a 2-parent household were only less likely to use than those in a 1-parent household. Notable gender differences were observed in the association of race with lifetime use. African American males and white females were more likely to have ever used marijuana. Females of Asian or “Other” race were less likely to have used.

Across gender, taking part in extracurricular activities was associated with lower rates of lifetime use, whereas having a job was associated with a greater likelihood of ever using. Furthermore, all risk behaviors were significantly associated with lifetime marijuana use. In all cases, participation in a risk behavior was associated with a greater likelihood of lifetime marijuana use. Data for demographic characteristics are listed in Table 1; data for psychosocial characteristics and risk behavior participation are in Table 2.

Demographic Characteristics of Participants Versus Lifetime and Past 30-Day Marijuana Use
Participant Psychosocial Characteristics and Risk Behavior Participation Versus Lifetime and Past 30-Day Marijuana Use

Characteristics Associated With Past 30-Day Marijuana Use, Stratified by Gender

As seen with lifetime marijuana use, grade in school, average grades, and family structure were significantly associated with past 30-day marijuana use across gender. For boys and girls, making an A/B average was associated with a lower risk of past 30-day marijuana use than was making a B/C average or Cs and lower. For boys, ninth graders were less likely to have used in the past 30 days than 12th graders, but for girls, ninth graders were less likely than 11th graders. As with lifetime marijuana use, males in a 2-parent household were less likely to have used in the past 30 days than males in either a 1-parent household or some other household, and females in a 2-parent household were only less likely than those in a 1-parent household. Although no race- or ethnicity-based differences were observed in males, white females were more likely to have used in the past 30 days, whereas African American and Asian/Other females were less likely.

Again, participation in extracurricular activities was protective against past 30-day marijuana use, and having a job was associated with a greater likelihood of use across genders. Also, participation in any risk behavior was associated with a greater likelihood of past 30-day marijuana use in both males and females.

Odds Ratios and Gender Interactions in Characteristics Associated With Lifetime Marijuana Use

Among characteristics associated with lifetime marijuana use, the highest AORs were seen for substance use: past 30-day smoking (boys = 16.05 and girls = 17.03), alcohol use (boys = 7.94 and girls = 8.99) and binge use (boys = 9.55 and girls = 8.83), and lifetime steroid use (boys = 5.96 and girls = 6.26). These were followed by other risk behaviors: carrying a weapon in the past 30 days (boys = 3.96 and girls = 4.92), past year fighting (boys = 3.73 and girls = 3.24), lifetime self-harm (boys = 2.56 and girls = 3.42), past year depressed mood with anhedonia (boys = 2.36 and girls = 1.78), and past year gambling (boys = 2.20 and girls = 1.66). Demographic and psychosocial characteristics were associated with both elevations in likelihood among male African Americans (AOR = 1.57) or male Hispanic/Latino (AOR = 1.37) and female white (AOR = 1.70) participants and were associated with protective effects for females of Asian or other descent (AOR = 0.67) and participation in extracurricular activities across genders (boys = 0.66 and girls = 0.37).

Five gender interactions were observed, with 3 based on race or ethnicity and 2 on psychosocial characteristics. African American males, but not females, were at greater odds of lifetime marijuana use (interaction P = 0.002); conversely, white females were at greater odds of lifetime use than were males (interaction P < 0.001). Being of Asian or other descent was protective for females, but it was not associated with male use (interaction P = 0.050). Furthermore, having a job elevated likelihood of lifetime use in females but not males (interaction P = 0.012), and it seemed that females who participated in extracurricular activities were less likely than males who participated to ever use marijuana (interaction P < 0.001). No risk behaviors evidenced an interaction with gender, though there was a trend (interaction P = 0.086) for males with past year depressed mood and anhedonia to have greater odds of past year marijuana use than females. All data on AORs and gender interactions are presented in Table 3.

Association Between Marijuana Use and Demographic, Psychosocial, and Risk Behavior Characteristics, Adjusting for Grade

Odds Ratios and Gender Interactions in Characteristics Associated With Past 30-Day Marijuana Use

A similar pattern of AORs emerged when examining past 30-day marijuana use. Substance users had the greatest likelihoods of marijuana use: past 30-day smokers (boys = 15.81 and girls = 10.86), alcohol users (boys = 11.85 and girls = 14.38) and binge users (boys = 13.18 and girls = 10.38), and lifetime steroid users (boys = 5.83 and girls = 7.38). The second highest set of AORs was seen with other risk behaviors: carrying a weapon in the past 30 days (boys = 3.57 and girls = 4.39), past year fighting (boys = 3.58 and girls = 3.09), lifetime self-harm (boys = 3.14 and girls = 2.88), past year depressed mood with anhedonia (boys = 2.27 and girls = 1.83), and past year gambling (boys = 2.52 and girls = 1.70). Having a job also was associated with an elevated likelihood of past 30-day marijuana use (boys = 1.34 and girls = 1.99), whereas participation in extracurricular activities was associated with a decreased likelihood of use (boys = 0.57 and girls = 0.40). No demographic characteristics in males were associated with an elevated AOR for past 30-day marijuana use, and in females, only white race was significantly associated (AOR = 1.92). Conversely, females of African American (AOR = 0.61) or of Asian/other descent (AOR = 0.59) were at lower odds of use.

Five gender-by-characteristic interactions were found for past 30-day marijuana use. For African American race, females were at lower odds of past 30-day use, but males had no significant association (interaction P = 0.020). Similarly, being of Asian/other descent was protective in females but not in males (interaction P = 0.012). However, females of white descent were at greater odds of past 30-day marijuana use than males (interaction P < 0.001). Participation in extracurricular activities was associated with lower odds of use in both genders, but females may derive greater benefit (interaction P = 0.032). Conversely, past 30-day cigarette use was associated with elevated AORs in both genders, but males seem to have greater elevations in odds than females (interaction P = 0.037).

Time From Initiation of Marijuana Use to Initiation of Regular Marijuana Use

Of participants, 500 males (26.2% of males) and 414 females (18.9% of females) endorsed regular marijuana use. The median age category for both initiation of use and initiation of regular use for males was at 13 or 14 years of age; for females, the median age of initiation was 13 or 14 years of age, but the median age of female initiation of regular use was between the 13 to 14 and 15 to 16 categories. Mann-Whitney U analysis indicated the presence of a gender difference in time of transition from initiation to regular marijuana use (Mann-Whitney U = 94,176.5, Z = −2.662, P = 0.008). Examination of the mean difference in transition indicated that males had a mean category difference of 0.53 from initiation to regular use (initiation mean = 4.49, regular use mean = 5.02), whereas females had a mean difference of 0.40 (initiation mean = 4.97, regular use mean = 5.37). Thus, it seems that females may have a quicker transition from initiation to regular marijuana use.


Before reviewing the results of this investigation, readers should keep in mind that a large sample size, such as this one, can lead to significant differences between groups, even with smaller differences between them. The most robust differences are found at more stringent significance levels, such as a P level ≤0.001; when significance was only achieved at a 0.05 or 0.01 level, this will be noted in the Discussion.

Overall, the primary findings of this study are that marijuana use is associated with a broad range of adverse health measures in both adolescent girls and boys, and that racial/ethnic and psychosocial characteristics are the main traits that evidence gender differences in association with both lifetime and past 30-day marijuana use in adolescents. For both the marijuana use variables, African American race, white race, Asian/other race, and participation in extracurricular activities significantly interacted with gender. The most robust of these interactions seemed to be in gender differences in lifetime marijuana use among whites, African Americans, and those participating in extracurricular activities (all P values <0.002); strong evidence was also found for gender differences in past 30-day marijuana use in white participants (P < 0.001). African American (P < 0.05) and Asian females seemed to be at lesser risk of marijuana use, whereas white females were at greater risk of use; females who participated in extracurricular activities seemed to derive greater protection from marijuana use than males who were involved in such activities. Also, having a job interacted with gender for lifetime marijuana use, with girls holding a job at greater risk for lifetime marijuana use (P < 0.05), and past 30-day cigarette use interacted with gender for past 30-day marijuana use, with male smokers at greater risk of marijuana use.

Data from this survey indicated that 40.4% of participants endorsed lifetime marijuana use and 24.5% endorsed past 30-day marijuana use. These rates are somewhat consistent with the findings of the MTF series of studies, though the 30-day prevalence rates of this sample appear higher. In 2005, the same year as this survey was administered, 44.8% of 12th grade students and 34.1% of 10th grade students endorsed lifetime marijuana use; 19.8% of 12th grade students and 15.4% of 10th grade students endorsed past 30-day use. Furthermore, when examining the correlates of lifetime and past 30-day marijuana use in adolescents, our results were also generally consistent with other studies (eg, Patton et al., 2005; Georgiades and Boyle, 2007; Medina et al., 2007; Wittchen et al., 2007; Brook et al., 2008), with all risk behaviors elevating the odds of concurrent marijuana use at a P value of <0.001, except for past year gambling in females (P < 0.01). The greatest elevations in odds were seen for the substance use variables, followed both the other risk behaviors. Demographic and psychosocial characteristics evidenced a more complex set of associations, with some traits elevating odds and some lowering odds. Finally, this study offers some evidence that female adolescents potentially had a faster transition from initiation of marijuana use to initiation of regular use.

It is somewhat difficult to compare our work with that of Tu et al. (2008), given that their focus was on frequency of use and our work focused on any lifetime or past 30-day use. Nonetheless, some potential areas of difference emerged. Tu et al. found that grade level was associated with frequent use in males only, whereas we found grade in school was associated with lifetime and past 30-day marijuana use across genders. Also, they found that poor self-reported mental health was associated with either frequent or heavy use in females only, whereas we found that past year depressed mood with anhedonia was associated with greater likelihood of marijuana use across genders. Furthermore, it seemed that males, not females, had greater odds of lifetime use when they had experienced past year depressed mood with anhedonia. It seems that the results of Tu et al. indicated only that certain subgroups of males were at greater risk for marijuana use, whereas we found that white females and those who have a job may be at greater risk as well.

Some of the differences in findings between this study and the work of Tu et al. may be explained by recent findings of gender differences in changes in rate of marijuana use among adolescents in Canada and the United States. In both nations, overall decreases in past year use were found, but only US males seemed to have significant declines in past year use during over the 4-year period from 2002 to 2006. Females did not evidence significant declines. In contrast, both male and female adolescents in Canada showed evidence of declines in past year marijuana use. Also, the sample examined by Tu et al. had higher 30-day marijuana use rates than this sample and included a significant subsample of Aboriginal participants (what would be called Native American or Alaskan Natives in the United States). These differences may have also contributed to the differences in findings. Future studies should attempt to clarify some of these potentially discrepant findings, perhaps through the inclusion of adolescents from different nations.

Five limitations of this study should be noted. First, 9.4% of participants did not have sufficient data for inclusion in these analyses. Excluded participants were more likely to be African American, Hispanic/Latino, male, and had higher rates of lifetime substance use. These missing data could have introduced some degree of selection bias, which must be considered when interpreting the results. Also, given the self-report nature of the survey, some misreporting by participants could have occurred, potentially altering the results. However, given the anonymous nature of the survey, any misreporting was likely to be small. Third, given the cross-sectional nature of the data, no causal inferences can be made. Future investigations should use longitudinal designs to allow for analyses of causality. Fourth, the use of categories to quantify age of initiation of marijuana use and age of initiation of regular use limits the conclusions that can be drawn from this data. Finally, although data were collected on family income, many participants endorsed “don't know,” and no data were collected on parental occupation or education level. This prevented us from examining the influence of socioeconomic status, which was a limitation. Future work should examine the importance of socioeconomic status for marijuana use.

These findings have specific implications for clinicians and those developing prevention and early intervention programs. First, it seems that the presence of other risk behaviors is the most important characteristic to screen for in determining which adolescents are most likely to be current marijuana users. Users of other substances are at a particularly elevated likelihood of marijuana use, which is consistent with previous work (eg, Wittchen et al., 2007).

Second, encouraging extracurricular activity participation seems to be an important protective factor for both male and female adolescents, with likelihood of marijuana use reduced by almost half in females and more than one quarter in males. In young adults, avoidance of boredom seems to be a prominent motive for marijuana use (Lee et al., 2007), and increasing prosocial activity engagement (for adolescents, this could include extracurricular activities) in adult substance users seems to be important for achieving abstinence (Rogers et al., 2008); thus, future research should examine whether participation in extracurricular activities functions to prevent or limit marijuana use by increasing community engagement and/or preventing boredom. Although this investigation found that extracurricular activity participation was associated with lower rates of marijuana use, the correlational nature of the research does not allow for causal inferences about the role of prosocial engagement by adolescents, including participation in extracurricular activities.

Third, it seems that race and ethnicity operate differentially in males and females, and screening should account for this. Female whites are at elevated likelihood of use, but females of African American (P < 0.05) or Asian/other descent seem to be protected; for males, African American or Hispanic/Latino ethnicity was associated with greater odds of past 30-day use (both P < 0.05). Examinations of adolescent substance users (including marijuana users) indicate that peer influence and support may operate differently by ethnicity (Wills et al., 2004), which may explain some difference found here, given the importance of and gender differences in peer influence for substance use (Rienzi et al., 1996). Other investigations indicate that levels of a variety of risk and protective factors differ by ethnicity in adolescents (Griffin et al., 2000), and the impact of specific factors such as parental involvement differ both by ethnicity and gender (Pilgrim et al., 2006). Finally, evidence suggests that cultural norms in Hispanic/Latino groups discourage female substance use (Canino, 1994); similar norms may deter use in African American females. Further investigation is needed, however, given that many of these results were found at less stringent (ie, P < 0.05) levels of significance.

In all, these results indicate that gender interacts with specific racial/ethnic and psychosocial characteristics to influence lifetime and past 30-day marijuana use in adolescents. Risk behaviors, with other substance use in particular, seem to be associated with the greatest elevations in odds of marijuana use across genders. Extracurricular activity participation seems to be the most robust factor associated with decreased odds of marijuana use, in both males in females. It is hoped that future investigations will expand on these findings, creating a clear profile of adolescent marijuana users that can be used to inform the development of prevention and early intervention programs.


Aldington S, Williams M, Nowitz M, et al. Effects of cannabis on pulmonary structure, function and symptoms. Thorax 2007;62:1058–1063.
Berthiller J, Straif K, Boniol M, et al. Cannabis smoking and risk of lung cancer in men: A pooled analysis of three studies in Maghreb. J Thorac Oncol 2008;3:1398–1403.
Brady KT, Randall CL. Gender differences in substance use disorders. Psychiatr Clin North Am 1999;22:241–252.
Bray JW, Zarkin GA, Ringwalt C, et al. The relationship between marijuana initiation and dropping out of high school. Health Econ 2000;9:9–18.
Bolla KI, Lesage SR, Gamaldo CE, et al. Sleep disturbance in heavy marijuana users. Sleep 2008;31:901–908.
Brook JS, Stimmel MA, Zhang C, et al. The association between earlier marijuana use and subsequent academic achievement and health problems: A longitudinal study. Am J Addict 2008;17:155–160.
Canino G. Alcohol use and misuse among Hispanic women: Selected factors, processes, and studies. Int J Addict 1994;29:1083–1100.
Di Forti M, Morrison PD, Butt A, et al. Cannabis use and psychiatric and cogitive disorders: The chicken or the egg? Curr Opin Psychiatry 2007;20:228–234.
Dorard G, Berthoz S, Phan O, et al. Affect dysregulation in cannabis abusers: A study in adolescents and young adults. Eur Child Adolesc Psychiatry 2008;17:274–282.
Fergusson DM, Boden JM, Horwood LJ. The developmental antecedents of illicit drug use: Evidence from a 25-year longitudinal study. Drug Alcohol Depend 2008;96:165–177.
Fried P, Watkinson B, James D, et al. Current and former marijuana use: Preliminary findings of a longitudinal study of effects on IQ in young adults. CMAJ 2002;166:887–891.
Georgiades K, Boyle MH. Adolescent tobacco and cannabis use: Young adult outcomes from the Ontario Child Health Study. J Child Psychol Psychiatry 2007;48:724–731.
Griffin KW, Scheier LM, Botvin GJ, et al. Ethnic and gender differences in psychosocial risk, protection, and adolescent alcohol use. Prev Sci 2000;1:199–212.
Guxens M, Nebot M, Ariza C. Age and sex differences in factors associated with the onset of cannabis use: A cohort study. Drug Alcohol Depend 2007;88:234–243.
Harvey MA, Sellman JD, Porter RJ, et al. The relationship between non-acute adolescent cannabis use and cognition. Drug Alcohol Rev 2007;26:309–319.
Johnston L, O'Malley PM, Bachman JG, et al. Monitoring the Future National Survey Results on Drug Use, 1975–2007. Volume I: Secondary School Students (NIH Publication No. 08-6418A). Bethesda, MD: National Institute on Drug Abuse, U.S. Dept. of Health and Human Services, National Institutes of Health; 2008.
Kandel DB, Chen K. Types of marijuana users by longitudinal course. J Stud Alcohol 2000;61:367–378.
Leatherdale ST, Hammond D, Ahmed R. Alcohol, marijuana, and tobacco use patterns among youth in Canada. Cancer Causes Control 2008;19:361–369.
Lee CM, Neighbors C, Woods BA. Marijuana motives: Young adults' reasons for using marijuana. Addict Behav 2007;32:1384–1394.
Medina KL, Nagel BJ, Park A, et al. Depressive symptoms in adolescents: Associations with white matter volume and marijuana use. J Child Psychol Psychiatry 2007;48:592–600.
Monshouwer K, VAN Dorsselaer S, Verdurmen J, et al. Cannabis use and mental health in secondary school children. Findings from a Dutch survey. Br J Psychiatry 2006;188:148–153.
Moore TH, Zammit S, Lingford-Hughes A, et al. Cannabis use and risk of psychotic or affective mental health outcomes: A systematic review. Lancet 2007;370:319–328.
Patton GC, Coffey C, Carlin JB, et al. Reverse gateways? Frequent cannabis use as a predictor of tobacco initiation and nicotine dependence. Addiction 2005;100:1518–1525.
Patton GC, Coffey C, Lynskey MT, et al. Trajectories of adolescent alcohol and cannabis use into young adulthood. Addiction 2007;102:607–615.
Pedersen W. Does cannabis use lead to depression and suicidal behaviours? A population-based longitudinal study. Acta Psychiatr Scand 2008;118:395–403.
Pedersen W, Mastekaasa A, Wichstrom L. Conduct problems and early cannabis initiation: A longitudinal study of gender differences. Addiction 2001;96:415–431.
Perkonigg A, Goodwin RD, Fiedler A, et al. The natural course of cannabis use, abuse and dependence during the first decades of life. Addiction 2008;103:439–449, discussion 450–431.
Pilgrim CC, Schulenberg JE, O'Malley PM, et al. Mediators and moderators of parental involvement on substance use: A national study of adolescents. Prev Sci 2006;7:75–89.
Reboussin BA, Hubbard S, Ialongo NS. Marijuana use patterns among African-American middle-school students: A longitudinal latent class regression analysis. Drug Alcohol Depend 2007;90:12–24.
Ridenour TA, Lanza ST, Donny EC, et al. Different lengths of times for progressions in adolescent substance involvement. Addict Behav 2006;31:962–983.
Rienzi BM, McMillin JD, Dickson CL, et al. Gender differences regarding peer influence and attitude toward substance abuse. J Drug Educ 1996;26:339–347.
Rogers RE, Higgins ST, Silverman K, et al. Abstinence-contingent reinforcement and engagement in non-drug-related activities among illicit drug abusers. Psychol Addict Behav 2008;22:544–550.
Schepis TS, Desai RA, Smith AE, et al. Impulsive sensation seeking, parental history of alcohol problems, and current alcohol and tobacco use in adolescents. J Addict Med 2008;2:185–193.
Schneider M. Puberty as a highly vulnerable developmental period for the consequences of cannabis exposure. Addict Biol 2008;13:253–263.
Substance Abuse and Mental Health Services Administration. Results From the 2007 National Survey on Drug Use and Health: National Findings, NSDUH Series H-34, DHHS Publication No. SMA 08-4343. Rockville, MD: Office of Applied Studies; 2008.
Tarter RE, Vanyukov M, Kirisci L, et al. Predictors of marijuana use in adolescents before and after licit drug use: Examination of the gateway hypothesis. Am J Psychiatry 2006;163:2134–2140.
Tu AW, Ratner PA, Johnson JL. Gender differences in the correlates of adolescents' cannabis use. Subst Use Misuse 2008;43:1438–1463.
Wagner FA, Anthony JC. Male-female differences in the risk of progression from first use to dependence upon cannabis, cocaine, and alcohol. Drug Alcohol Depend 2007;86:191–198.
Walitzer KS, Dearing RL. Gender differences in alcohol and substance use relapse. Clin Psychol Rev 2006;26:128–148.
Wills TA, Resko JA, Ainette MG, et al. Role of parent support and peer support in adolescent substance use: A test of mediated effects. Psychol Addict Behav 2004;18:122–134.
Wittchen HU, Frohlich C, Behrendt S, et al. Cannabis use and cannabis use disorders and their relationship to mental disorders: A 10-year prospective-longitudinal community study in adolescents. Drug Alcohol Depend 2007;88(suppl 1):S60–S70.

cannabis; adolescent; gender differences; risk behaviors

© 2011 American Society of Addiction Medicine